Biology

Biology fields

Okayama University Year 2007

Tissue-specific interactions of TNI

isoforms with other TN subunits and

tropomyosins in C. elegans: The role of

the C- and N-terminal extensions

Md. Ziaul Amin∗ Tetsuya Bando† Razia Ruksana‡

Frederick Anokye-Danso∗∗ Yasuo Takashima††

Yasuji Sakube‡‡ Hiroaki Kagawa§

∗Okayama University†Okayama University‡Okayama University∗∗Okayama University††Okayama University‡‡Okayama University§Okayama University, hkagawa@cc.okayama-u.ac.jp

This paper is posted at eScholarship@OUDIR : Okayama University Digital InformationRepository.

http://escholarship.lib.okayama-u.ac.jp/biology general/35

Tissue-specific interactions of TNI isoforms with other TN subunits and tropomyosins in C. elegans: The role of the C- and N-terminal extensions

Md. Ziaul Amin, Tetsuya Bando, Razia Ruksana, Frederick Anokye-Danso, Yasuo Takashima,

Yasuji Sakube, and Hiroaki Kagawa*

Division of Bioscience, Graduate School of Natural Science and Technology, Okayama University,

Okayama 700-8530, Japan

*Corresponding author

Tel: 81-86-251-7865; Fax: 81-86-251-7876

E-mail address: hkagawa@cc.okayama-u.ac.jp

Running title: Interactions of troponin I isoforms in C. elegans

Keywords: Troponin I; Caenorhabditis elegans; Troponin T; Troponin C; Tropomyosin;

C-terminal extension

- 2 -

Abstract

The aim of this study is to investigate the function of the C-terminal extension of three troponin I

isoforms, that are unique to the body wall muscles of Caenorhabditis elegans and to understand the

molecular interactions within the TN complex between troponin I with troponin C/T, and

tropomyosin. We constructed several expression vectors to generate recombinant proteins of three

body wall and one pharyngeal troponin I isoforms in Escherichia coli. Protein overlay assays and

Western blot analyses were performed using antibodies. We demonstrated that pharyngeal TNI-4

interacted with only the pharyngeal isoforms of troponin C/T and tropomyosin. In contrast, the

body wall TNI-2 bound both the body wall and pharyngeal isoforms of these components.

Similar to other invertebrates, the N-terminus of troponin I contributes to interactions with troponin

C. Full-length troponin I was essential for interactions with tropomyosin isoforms. Deletion of the

C-terminal extension had no direct effect on the binding of the body wall troponin I to other muscle

thin filament troponin C/T and tropomyosin isoforms.

Abbreviations used: TN, troponin; TNI, troponin I; TNC, troponin C; TNT, troponin T; TM,

tropomyosin

- 3 -

1. Introduction

Muscle contraction is the result of a series of protein-protein interactions. In striated muscle,

the thin filament complex of troponin (TN) and tropomyosin (TM) regulates contraction. The

troponin complex components include the calcium-binding protein, troponin C (TNC), the

tropomyosin (TM)-binding protein, troponin T (TNT) and troponin I (TNI) which is involved in

inhibition of the actomyosin ATPase activity [1, 2]. Recent studies on crystals of the core TN

complex suggest that in the absence of calcium, TNI-binding to actin holds the TN-TM complex in

a “closed” state thereby preventing myosin from binding to actin [3, 4].

Calcium ions (Ca2+) released into the muscle cells following neural stimulation bind to

TNC, which undergoes a conformation change that alters its relationship with TNI, resulting in the

release of TNI-binding to actin. Biochemical and physicochemical studies of muscle contraction

have been carried out using proteins from different tissues of vertebrates. Multiple forms of troponin

subunits have been identified in cardiac [5], slow skeletal [6] and fast skeletal muscles [7]. The

detection of TNI isoforms by specific antibody staining provides a convenient and reliable method

of muscle cell typing and the presence of TNI in serum can be used for the detection of disease in

skeletal and cardiac muscles [8]. The immunological detection of serum cardiac TNI is widely used

in cardiology as an index of myocardial damage [9]. Slow and fast isoforms are present in the same

skeletal muscle cell as a consequence of cross innervations, hormone intervention, and in

pathological conditions [10]. Thus, specific TNI isoforms probably play important roles in

determining the distinctive functional properties of various muscle types from different animals.

Specific amino acid sequences in muscle proteins have been shown to interact with specific

sites. The first report on TNI fragments that bind TNC was done by Syska et al. [11]. They report

that TNI fragment of the rabbit skeletal containing residues 1-21 and 96-116 bind TNC, and the

latter also bind to actin. It has also been summarized in the rabbit skeletal TNT that the N-terminal

residues 160-260 bound to TNC, and the C-terminal residues 216-263 might be the TNI interaction

site[12]. Thus, specific interaction sites in the components form part of a functional complex.

Comparisons of these amino acid sequences reveal that a high sequence homology of the

TNC-binding site is present towards the N-terminal region of TNI isoforms containing N-terminal

extensions that are seen in invertebrates and vertebrate hearts. It has been reported that

bisphosphorylation of the N-terminal extension of human cardiac TNI alters the Ca2+-binding

properties to TNC [13]. Similarly the high sequence homology of the actin/TNC-binding site

exists within the central portions of TNI isoforms from many animals. These results of biochemical

- 4 -

studies were confirmed when atomic structures of the core TN complex recently became available

[3, 4]. Probably, the regulatory complex of TN components shares common structural bases within

the core complex but with some variation depending on muscle or animal type. Although many

detailed analyses of interaction sites between the TN components still continues, few report on the

overall interactions between regulatory components and include all isoforms derived from a single

genome.

TNI isoforms are expressed in two patterns: tissue specific and developmentally

regulated. Previously, we have shown that three genes, namely, tni-1, tni-2/unc-27 and tni-3 are

expressed in the body wall muscles, and that tni-4 exhibits pharyngeal-specific expression in

Caenorhabditis elegans [14]. An approach to understand the role of TNI in the regulation of muscle

contraction is to study the structural and functional differences between the various TNI isoforms.

In our previous study, we also compared the amino acid sequences of four TNI isoforms of C.

elegans with those of Drosophila, crayfish, and the cardiac, slow skeletal and fast skeletal muscles

of rabbit [14]. Comparisons of these amino acid sequences reveal that a high sequence homology

of the TNC-binding and the actin/TNC-binding sites exists in all C. elegans TNI isoforms.

Interestingly, three body wall type TNI-1, TNI-2 and TNI-3 isoforms of C. elegans contain unique

C-terminal extensions. C-terminal extensions containing approximately 20 glutamic acid residues

are common in all body wall TNI isoforms but not in pharyngeal TNI isoform of C. elegans, the

TNI isoforms of Drosophila, crayfish and rabbit [14]. The biological importance of the C-terminal

extension remains unknown structurally and functionally.

In this study, we focused on the importance of the C-terminal extensions of three body wall

TNI isoforms as well as the tissue-specific interaction of TNI isoforms with other thin filament

components. To investigate the interactions of the TNI isoforms and those fragments with other

thin filament proteins, we constructed recombinant proteins of the N-, C-, and C-terminal

extension-deletion fragments of TNI isoforms and performed Western blot analysis using

antibodies. The C-terminal extensions had no direct role in the TNC, TNT and TM interactions.

Molecular binding assays of TNI with other thin filament components provide an insight into how

thin filament regulatory proteins evolved to specialize in their interactions to regulate muscle

contractions. This is the first report on how the components of the TN complex of C. elegans

interact with other thin filament components.

- 5 -

2. Materials and methods

2. 1. Construction of expression vectors

2. 1. 1. Construction of the N-terminal, C-terminal and C-terminal extension-deletions of the TNI

expression vectors

Expression vectors of the N- and C-terminal fragments of TNI-2, pCTNI-21–117 and

pCTNI-2118–242 were constructed by inserting BamHI restriction fragments corresponding to a 354

bp 5'-region and a 378 bp 3'-region of the tni-2 cDNA [14] into the BamHI site of pET-28a(+)

vector, respectively. These restriction fragments were prepared by PCR amplification using tni-2

cDNA as a template with the following sets of primers followed by BamHI digestion; TNI-2N-s

(5'-AAG GAT CCA TGA GTG AAG AAG CCG GAG A-3') and TNI-2N-as (5'-AAG GAT CCG

TTG ATG TCG TAC TTC TCC TCC-3') primers for pCTNI-21–117, and TNI-2C-s (5'-AAG GAT

CCA TCA ACT ACG TCG TCT CCC AGA-3') and TNI-2C-as (5'-AAG GAT CCT GGA TTT

AGG AGT TTA CTC CTC-3') primers for pCTNI-2118–242. pCTNI-294–242 was constructed by

inserting a XhoI-HindIII restriction fragment corresponding to the 453 bp 3'-region of the tni-2

cDNA into the XhoI-HindIII site of pET-28c(+) vector. The C-terminal extension-deletion

construct of TNI-2 expression vector pCTNI-21-220 was constructed by inserting a BamHI

restriction fragment corresponding to the 5'-region of the tni-2 cDNA, which is derived from the

subcloned vector pGEX-CTNI-2 [14], into the BamHI site of the pET-28a(+) vector. This

restriction fragment was prepared by PCR using pGEX-CTNI-2 as a template with pGEX5'

(5'-GGG CTG GCA AGC CAC GTT TGG TG-3') and UNC-27-P2 (5'-TTG GAT CCT TAC

TCA GCT TCT GGA GCA ACG-3') primers, followed by BamHI digestion. pCTNI-41-166 was

constructed by inserting an EcoRI restriction fragment corresponding to the 530 bp 5'-half region of

the tni-4 cDNA from yk328a10 into the EcoRI site of the pET-28b(+) vector. This restriction

fragment was prepared by PCR using yk328a10 as a template with M13 (–21) and

TNI-4-s-Eco-pETb (5'-GCG AAT TCA ACC ATG AGT GAC GTT GAC G-3') primers. Note that

the TNI-4-s-Eco-pETb primer altered the stop codon in the 5' untranslated region (UTR) of the

tni-4 cDNA. pCTNI-491–194 was constructed by inserting an XhoI restriction fragment

corresponding to the 519 bp 3'-region of the tni-4 cDNA from yk328a10 into the XhoI site of the

pET-28b(+) vector.

Expression vectors of other TNI-1/-3 isoforms were also constructed briefly as follows.

pCTNI-11-126 was constructed by inserting an EcoRI restriction fragment of tni-1 cDNA from

- 6 -

yk103h4 into the EcoRI site of pET-28b(+) vector. pCTNI-1106-250 was constructed by inserting the

XhoI restriction fragment of tni-1 cDNA from yk103h4 into the XhoI site of pET-28b(+) vector.

pCTNI-31-125 was constructed by inserting an EcoRI restriction fragment of tni-3 cDNA from

yk338a12 into EcoRI site of pET-28a(+) vector. pCTNI-3118-260 was constructed by inserting the

BglII-XhoI restriction fragment of tni-3 cDNA from yk338a12 into BglII-XhoI site of pET-28c(+)

vector. In summary, the N-terminal, C-terminal and C-terminal extension-deletion constructs of the

TNI isoforms, as shown by the bars in Fig. 1, were cloned into the pET-28a(+), pET-28b(+) and

pET-28c(+) vectors (Novagen), respectively and each of fragments were produced in E. coli strain

BL21(DE3).

2. 1. 2. Construction of the TNT-1, TNT-4, TNC and TM expression vectors

pCTNT-1 was constructed by inserting an EcoRI restriction fragment corresponding to

the 1215-bp full-length tnt-1/mup-2 cDNA from yk1354g3 into the EcoRI site of the pET-28a(+)

vector. This restriction fragment was prepared by PCR using yk1354g3 as a template with TNT-1-s

(5'-TTG AAT TCA TGT CCG ACG AGG AGG AGG TAT AC-3') and TNT-1-as (5'-TTG AAT

TCA TTC GAC AAC GAC CTC TTC TCC-3') primers, followed by EcoRI digestion. pCTNT-4

was constructed by inserting a BamHI restriction fragment corresponding to the 1044-bp

full-length tnt-4 cDNA from yk1174g10 into the BamHI site of the pET-28a(+) vector. This

restriction fragment was prepared by PCR using yk1174g10 as a template with TNT-4-s (5'-CCT

GGA TCC TCG ATG AAC ATG TCT GAC GAG GAA TAC TCC G-3') and TNT-4-as (5'-TCT

GGA TCC TTA ATA GTC TTC CTC TTC CTC GGC-3') primers. The plasmid vectors for the

bacterial expression of TNC-1 and TNC-2 have been described previously [14, 15].

Some expression vectors of TNC, TNI and tropomyosin of C. elegans were used as

reported previously (Table 1) [14-16]. TMI and TMIII expression vectors and the cDNA clones of

CeTMI and CeTMIII were used as described previously [16]. All sub-cloned constructs were

verified by DNA sequencing.

2. 2. Western blot analysis and protein overlay assay

Western blot analysis was performed using the ECL detection system (Amersham, now

GE Healthcare Bio-Sciences) by using the following specific antibodies: anti-CeTMIII [16],

anti-CeTNC-1/-2 [15] and anti-CeTNI-2/-4 [14]. Table 1 also describes the antibodies used in this

study. Anti-CeTNC-1 and anti-CeTNC-2 were tissue-specific as they detected only body wall

- 7 -

TNC-1 and pharyngeal TNC-2, respectively (Fig. 2c) [14]. Anti-CeTMIII detected both the body

wall TMI and the pharyngeal TMIII isoforms (Fig. 2d). Cross-reactivity of the TMI isoform with

anti-CeTMIII, which was raised to detect the pharyngeal tropomyosin, TMIII isoforms, is due to

the sharing of some amino acid residues from common exons [16, 17].

The body wall TNI-2 and TMI, and the pharyngeal TNI-4 and TMIII proteins of C.

elegans were produced in E. coli strain JM109, which were transformed with each of cDNA

expression vectors. cDNA clones of TNI-2/-4 and TMI/III were cloned into pGEX-4T series and

pUR289, respectively [14,16]. The TNC-1 and TNC-2 proteins were produced in E. coli strain

BL21 (DE3) transformed with the pCTNC-1 and pCTNC-2 vectors, respectively [15]. Note

that the expressed proteins had no tag. TNT-1 and TNT-4 proteins were produced in E. coli BL21

(DE3) transformed with the pCTNT-1 and pCTNT-4 vectors, which expressed His-tagged TNT-1,

the body wall isoform, and TNT-4, the pharyngeal TNT isoform of C. elegans. Protein samples

were prepared according to the standard procedure in which 0.05 µg/ml TNI, 0.02 µg/ml TNC,

0.04 µg/ml tropomyosin and 0.05 µg/ml TNT were run on SDS-PAGE and blotted onto a nylon

membrane. The processed membranes were immersed in 20 µg/ml TNI, 30 µg/ml TNC, 25 µg/ml

tropomyosin and 50 mM Tris-HCl (pH 6.8) solution with 1 mM CaCl2 or 2 mM EGTA at 4 °C for

12 hours. After washing, Western blot analysis was performed using 0.05 mg/ml rabbit polyclonal

anti-CeTNI-2, anti-CeTNI-4, anti-CeTNC-1 and anti-CeTNC-2 antibodies, anti-CeTMIII

anti-serum were used as the primary antibody, and 0.2 mg/ml of horseradish anti-rabbit IgG

(Amersham Biosciences) was used as the secondary antibody; the samples were incubated at room

temperature for 1 hour and analyzed by using the ECL detection system.

- 8 -

3. Results

3. 1. Characterization of the fragments of the TNI, TNT and TM isoforms and their antigenic

characters

In C. elegans all isoforms of any muscle protein are identified and are available from the

database, WormBase. We expressed several muscle thin filament proteins in E. coli by constructing

expression vectors that are summarized in Table 1. We also constructed expression vectors to

produce the N- and C-terminal fragments of three body wall isoforms (TNI-1, TNI-2 and TNI-3)

and one pharyngeal isoform (TNI-4) (see Materials and methods; Fig. 1 and Table 1). The

fragments consisted of two parts: the N-terminal fragment containing the TNC-binding site and the

C-terminal fragment containing the actin/TNC-binding site. For example, deletion of the

C-terminal extension fragment of TNI-2 was designated as TNI-21–220. The molecular sizes of the

N-terminal TNI-21–117, C-terminal TNI-294–242, TNI-2118–242 and C-terminal extension- deletion

TNI-21–220 fragments of the TNI-2 isoform were confirmed by SDS-PAGE (data not shown) and

are summarized in Table 1. The products of these fragments were detected by Western blot analysis

by using the anti-CeTNI-2 antibody that stained all four TNI isoforms of C. elegans (Fig. 2a) [14].

Size differences were observed in the fragments of the TNI isoforms due to the design of the

expression vectors. We focused on the body wall TNI-2 isoform, a major known TNI component

of the body wall muscle and the product of the tni-2/unc-27gene [14].

The expression vectors pCTNT-1 and pCTNT-4 expressed body wall TNT-1, a product of

mup-2 [18] and pharyngeal TNT-4, a product of tnt-4, respectively. The molecular sizes of these

recombinant proteins are summarized in Table 1. These proteins and other thin filament

components were used in the molecular interaction experiments. The anti-CeTNI-2 antibody,

which was raised against the major body wall type, cross-reacted with all TNI isoforms, three body

wall types and one pharynx type (Fig. 2a). This antibody also detected the N- and C-terminal

fragments and the full length of all TNI isoforms; which suggested the presence of more than one

recognition site in the TNI-2 isoform (Fig. 2a). As shown in Fig. 2b, the anti-CeTNI-4 antibody

specifically detected the pharyngeal isoform, TNI-4. Fig. 2c shows that antisera against each of

both the body wall type TNC-1 and the pharynx type TNC-2 isoforms had tissue specificity [14,

15]. The recombinant proteins and antibodies produced in this study are summarized in Table 1 and

Materials and methods.

- 9 -

3. 2. Molecular interactions of body wall and pharyngeal TNI isoforms with other thin filament

components

3. 2. 1. Interaction between TNI and TNC isoforms

Both the TNC-1 and TNC-2 isoforms were overlaid on blotted N-terminal TNI-21–117,

C-terminal TNI-294–242 and TNI-2118–242, and the C-terminal extension-deleted TNI-21–220 fragments

in the presence or absence of Ca2+. Similar overlay experiments were also performed with the TNC

isoforms and the pharyngeal TNI-41–166 and TNI-491–194 fragments. Each of the overlaid TNC

isoforms was detected by using the anti-CeTNC-1 or anti-CeTNC-2 antibody (Fig. 3a-b). These

results indicated that body wall TNC-1 bound specifically with the N-terminal body wall TNI-21–117

fragment, but not with the pharyngeal TNI-4 isoform (Fig. 3a, upper and lower panels) [14].

Pharyngeal TNC-2 interacted with both the N-terminal TNI-21–117 and TNI-41–166 fragments and

also the full-length TNI-2 and TNI-4 isoforms (Fig. 3b, upper and lower panels) [14]. The

C-terminal extension-deleted TNI-21–220 fragment of body wall TNI-2 isoform was able to bind

with both the TNC isoforms but both the C-terminal body wall TNI-294–242, TNI-2118–242 and

pharyngeal TNI-491–194 fragments did not bind with either the body wall TNC-1 or pharyngeal

TNC-2 isoforms, respectively (Fig. 3a-b). All interaction profiles between each pair of TNI and

TNC isoforms are summarized in Fig. 3c. We concluded that the N-terminal body wall TNI-2

interacted with both the body wall and pharyngeal TNC isoforms, but the N-terminal pharyngeal

TNI-4 interacted only with pharyngeal TNC-2.

The extent of the interaction varied among the TNI isoforms and the N-terminal fragment

used depending on the presence or absence of Ca2+. However, it was difficult to determine whether

these results reflect quantitative differences in these interactions. We hereafter focus on the TNI-2

isoform to study the molecular interactions with other thin filament components from both the

body wall and the pharynx.

3. 2. 2. Interaction between TNI and TNT isoforms

We performed protein overlay assays on full-length and fragmented body wall TNI

isoforms with both body wall and pharyngeal TNT isoforms (Fig. 4). Full length TNI-2, the

N-terminal TNI-21–117, the C-terminal TNI-294–242 and the C-terminal extension-deleted TNI-21–220

fragments interacted with both the TNT isoforms (Fig. 4a, b, d and e) but the C-terminal

TNI-2118–242 fragment did not (Fig. 4c).

The full length TNI-4 and two fragments TNI-41–166 and TNI-491–194 bound only the

- 10 -

pharyngeal TNT-4 isoform (Fig. 4f–h). The results of the interaction of the TNI isoforms with the

TNT isoforms in the presence or absence of Ca2+ were almost similar (data not shown). The

interaction profiles between the TNI and TNT isoforms are summarized in Fig. 4i. In conclusion,

the body wall TNI isoforms interacted with both the body wall TNT-1 and the pharyngeal TNT-4

isoforms, but the pharyngeal TNI-4 only bound the pharyngeal TNT-4. The TNT interaction sites

of four TNI isoforms in Fig. 1 were presented more details in Fig. 6a (see Discussion).

3. 2. 3. Interaction between TNI and TM isoforms

We further determined which of the TNI terminal regions interact with the tropomyosin

isoforms of C. elegans. The pharyngeal TNI-4 isoform bound only to the pharyngeal tropomyosin,

TMIII (Fig. 5b). Interestingly, the TNI-21–117, TNI-294–242, TNI-2118–242, TNI-41–166 and TNI-491–194

fragments had lost the ability of the body wall and pharyngeal TNI isoforms to interact with

tropomyosin isoforms (Fig. 5a-b). We conclude that body wall TNI-2 bound TNC/T and TM

components from both tissues but pharyngeal TNI-4 interacted only with the pharyngeal

components (Fig. 1). All the tissue-specific interactions of the TNI isoforms with other thin filament

components are summarized in Fig. 1 (right panel) and are schematically shown in Fig. 6b (see

Discussion).

- 11 -

4. Discussion

By performing the protein overlay assays using antibodies as probes we have determined

the molecular interactions of the body wall TNI isoforms with other thin filament components,

TNC, TNT and TM. We obtained evidence that the pharyngeal TNI-4 isoform bound

tissue-specifically with other thin filament components but that the body wall TNI isoforms

interacted with these components from both tissues. The functional significance of the C-terminal

extension, which is unique to the body wall TNI isoforms of C. elegans, was not found by this

approach.

4. 1. Molecular interactions among troponin components and tropomyosin

4. 1. 1. Function of the C-terminal extension in the body wall TNI-2 isoforms

In this study, we found that the N-terminal TNI-21–117, the C-terminal TNI-294–242,

C-terminal extension-deleted TNI-21–220 and full-length TNI-2 all bound in a similar manner with

both the body wall and pharyngeal TNT isoforms (Fig. 4a, b, d and e). The evidence obtained in

this in vitro study confirmed that there was no difference in the interactions of body wall TNI

isoforms with or without the C-terminal extension (Fig. 4d and e) and in the presence or absence of

Ca2+ (data not shown). Similarly, the overlay assay results of TNC and TM isoforms also show that

the C-terminal extension-deleted TNI-21–220 fragment is able to bind to both TNC and TM isoforms

(Fig. 3a-b and Fig. 5a-b). Therefore, these results imply that the C-terminal extension has no direct

role on binding to TNC, TNT and TM isoforms in C. elegans. Although we did not study the

molecular interactions of the C-terminal TNI extension deletions of the two other body wall

isoforms, the sequence similarity of the three TNI isoforms suggests that the C-terminal extension

could have functions that were not detected by in vitro overlay assay. As the atomic structure of the

core component of the human cardiac troponin complex does not include the TNI C-terminal

region of this end of TNI [4], it could still have an important role in actin-myosin interactions.

Tanaka et al. [19] reported that the ATNI232-292 C-terminal fragment of Akazara scallop TNI

inhibited actomyosin-tropomyosin Mg-ATPase activity. Al-Hillawi et al. [20] also reported that the

recombinant human cardiac TNI inhibited the acto-S1 Mg-ATPase activity. This inhibition was

potentiated by the presence of tropomyosin and was reversed by the addition of the TNC to the

system. Although we have not done experiments on the C-terminal extension to determine

functional effects on ATPase activity, we assume that the C-terminal extension of the three body

- 12 -

wall TNI isoforms of C. elegans has some biological function.

In muscle studies for C. elegans system [21], in vivo experiments could be used to

determine the functional importance of the C-terminal extension. This might be achieved by

studying whether the pharyngeal or C-terminal extension-deleted body wall TNI isoform could

rescue the mutant phenotype of tni-2/unc-27 animals that have a mutation in the major body wall

isoform gene, tni-2 [14].

4. 1. 2. Molecular interactions between the TNI and TNT isoforms

Immunoblotting, cross-linking and competitive labeling studies on the rabbit fast skeletal

TNI and TNT suggest that the TNI 58-107 segment bound to TNT [22-24]. The atomic structure of

the human cardiac troponin core complex shows that the helical part of H2 (I) interacted to H2 (T2)

[4]. This region has been designated to the IT-arm [4]. We have focused on the IT-arm region,

residue F90-R136 in human cardiac, of TNI in comparing residues among the four TNI isoforms

of C. elegans and other animals (Fig. 6a) [14]. Our overlay assay results of the N-terminal

TNI-21-117 and the C-terminal TNI-294-242 fragments of TNI-2 isoform suggest that TNI and TNT

interaction sites are located in residues 94-117 (Fig. 1, Fig. 6a). The sequence alignment for the

interaction sites in the four TNI isoforms of C. elegans (shaded in Fig. 6a) show that the interaction

sites located in the N-terminal part of the IT-arms consist of three helical surfaces, three a and d

surface of the heptad. In contrast to that, the pharyngeal TNI-4 isoform bound only with the

pharyngeal TNT-4 isoform (Fig. 4f-h), whereas the body wall TNI-2 interacted with both tissue

TNT isoforms (Fig. 4a, b, d, e and i). This suggests that the sites at which TNI and TNT interaction

occurred contain some differences. From these results we assume that functional differences

between the body wall and pharyngeal TNI isoforms came from deleting three amino acid residues

in the pharyngeal isoform, TNI-4. The sequence alignment and the secondary structure prediction

algorithm also suggest that this region can form a helical configuration. In the outside of the

N-terminal of the helix, one proline residue is present in both TNI-1 and TNI-2 but two proline

residues occur in TNI-3 and TNI-4. These will cause tilting of the helical rod of the IT-arm.

Compared to the actin/TNC-binding sites of these TNI isoforms, the identity of TNT-binding sites

among animals does not have a high sequence homology but there are some sequences homologies

for forming helical conformations. Functional differences in the TNT-binding site between these

animals may arise from fine-tuning the conformation of these regions between TNI and TNT.

More detailed experimental studies on functional differences will be needed to investigate this

further.

- 13 -

4. 1. 3. Molecular interactions of body wall TNI-2 with other thin filament components

Our results that the pharyngeal TNI-4 only bound with the pharyngeal TNC-2 isoform but

not with body wall TNC-1 (Fig. 3a-b) are similar to those in a previous report that the nature of the

TNI and TNC interaction is very different in different organisms and tissues [19]. The result of our

overlay assays of TNC binding to TNI isoforms implies that the N-terminal regions of both the

TNI-2 and TNI-4 isoforms interacted with pharyngeal TNC-2 (Fig. 3a-b). Farah et al. [25] also

reported that the N-terminal region of chicken skeletal muscle TNI bound to the C domain of TNC.

The NMR structure of cardiac TNI reveals that the N-terminal fragments of cardiac TNI interact

with the C-terminal region of cardiac TNC [2]. Therefore, our results suggest that the importance of

the N-terminal region of TNI in interactions with TNC isoforms could be common in both

vertebrates and invertebrates. The reason why all C-terminal fragments of TNI did not to bind to

TNC although they all contain actin/TNC-binding region could be the character of invertebrates.

This was consistent to the case that molluskan Akazara scallop TNI232-292, which contained the

actin/TNC-binding region, did not interact with TNC even in the presence of Ca2+ [19]. It is of

interest to study how molecular interactions among TN complex are developed during evolution.

Figure 5c summarizes the interactions between the two tissue-specific TNI and

tropomyosin isoforms. These results suggest that the conformation of the TNI isoform altered after

fragmentation, as the fragmented forms could not bind the tropomyosin isoforms (Fig. 5a-b).

Similar results were observed in overlay assays of the tropomyosin and TNI isoforms in the

absence of Ca2+ (data not shown). It has been reported that truncated troponin binds weakly to

tropomyosin, regardless of Ca2+ presence [26]. Our results are consistent with this. It has also been

reported that TNI binds weakly to tropomyosin in the presence or absence of Ca2+, whereas the

TNI/TNC complex shows no detectable interaction with tropomyosin under similar conditions [27].

Thus, the direct binding of TNI with tropomyosin could be important in the regulation of muscle

contraction.

4. 2. Possible limitation of the protein overlay assay

Although the yeast two-hybrid system is a well-characterized method for analyzing

protein-protein interaction [28], it can only identify potential binding partners for a protein of

interest; no information is provided about whether an interaction actually occurs in vivo or what the

consequences of binding might be. Pull-down approaches [29] can reveal protein associations that

- 14 -

occur in a cell of interest, but still cannot address the functional consequences of those interactions.

Thus, it remains difficult to identify those specific protein interactions that are functionally relevant

to a biological activity of interest. Here, we show that protein overlay assays can be used to study

molecular interactions between TNI and other thin filament components. This approach is more

convenient than column chromatography. It is quicker and uses less equipment and protein. The

degree of protein refolding, the antibody used as probe, and the order of overlaid or blotted proteins

are important factors for success. Helical structures especially are easily recovered in the original

structure and can interact with other proteins under the experimental conditions. Our results on

Western analyses show that isoform specificities of each fragmented TNI isoform are detected by

the corresponding antisera (Fig. 2). Depending on solubility or the size of proteins the interaction

can vary. We designed our experiments so that the soluble and smaller proteins were overlaid onto

blots of the less soluble or larger protein. Unfortunately we could not detect differences in protein

binding in the presence or absence of Ca2+. It is difficult to control the quantities of the sample

solution. Despite these limitations the overlay approach is a powerful means to determine the

overall relationships between wide ranges of proteins from different sources. We have discussed in

more details our use of this approach to study TNI isoforms and other thin filament components.

4. 3. Gene duplication of thin filament components

Based on the assay results of the interaction between TNI and TNC (Fig. 3), TNI and TNT

(Fig. 4) and TNI and TM (Fig. 5) in this study, we summarized the molecular interactions between

these proteins in two tissues (Fig. 6b). Further, we observed the important fact that pharyngeal

TNI-4 interacted only with other pharyngeal thin filament proteins, and not with the body wall

proteins (Fig. 6b). We assume that this was due to evolutionary and functional reasons: Thin

filament proteins expressed in the pharynx interacted with each other, as was observed in the

overlay assay between the pharyngeal isoform pairs (step 1). In evolution, the body wall TNI gene,

probably tni-3, duplicated subsequently, from tni-4 (step 2) [14]. Two other body wall TNI genes

are likely duplicates of the original body wall TNI gene tni-3; this is also implied their tandem array

on chromosome X. However, these genes do not differ considerably in their functions, although we

have not determined this experimentally. These three body wall TNI isoforms can still interact with

the pharyngeal isoforms (step 3). The body wall type genes of the thin filament components TNC-1,

TNT-1 and tropomyosin TMI/II subsequently duplicated together with a mutation in the

corresponding pharyngeal genes (step 4). The products of these genes could not interact with

- 15 -

pharyngeal TNI-4 because of the mutations that had occurred following gene duplication. Finally,

the three body wall TNI isoforms could interact with the body wall TNC, TNT and TM isoforms

(step-5). Troponin gene duplication during evolution and current isoform interactions can be

explained in this manner. The sequence homology of the TNT-binding sites in the four TNI

isoforms was lower than those of TNC- and actin/TNC-binding sites (Fig. 6a) [14]. Even with

regard to the helical configuration of the TNI-TNT interaction site, the functional differences that

depend on the type of amino acid residues of the isoforms could be important, although we have

yet to determine any functional differences between these body wall isoforms. The hypothesis

presented in this study will be useful to explore further the interactions within the troponin complex

and other thin filament components in detail.

- 16 -

Acknowledgements

We thank Dr. Yuji Kohara for providing the yk series of cDNA clones and all laboratory

members for their encouragement and technical support. We acknowledge Prof. John C. Sparrow

for his generous help in critically reading the manuscript. Md. Z. A. is a foreign student who

received a student scholarship from the Ministry of Education, Culture, Sports, Science and

Technology of Japan. This work was supported by a grant to H. K. from the Ministry of Education,

Culture, Sports, Science and Technology of Japan.

- 17 -

References

[1] I. Ohtsuki, K. Maruyama, S. Ebashi, Regulatory and cytoskeletal proteins of vertebrate skeletal

muscle, Advan. Protein Chem. 38 (1986) 1-67.

[2] A.M. Gordon, E. Homsher, M. Regnier, Regulation of contraction in striated muscle, Physiol. Rev.

80 (2000) 853-924.

[3] D.G. Vassylyev, S. Takeda, S. Wakatsuki, K. Maeda, Y. Maeda, Crystal structure of troponin C in

complex with troponin I fragment at 2.3-A° resolution, Proc. Natl. Acad. Sci. U.S.A. 95 (1998)

4847-4852.

[4] S. Takeda, A. Yamashita, K. Maeda, Y. Maeda, Structure of the core domain of human cardiac

troponin in the Ca2+-saturated form, Nature 424 (2003) 35-41.

[5] R.J.A. Grand, J.M. Wilkinson, The amino acid sequence of rabbit cardiac troponin I, Biochem. J.

159 (1976) 633-641.

[6] R.J.A. Grand, J.M. Wilkinson, The amino acid sequence of rabbit slow-muscle troponin I, Biochem.

J. 167 (1977) 183-192.

[7] Z. Sheng, B.S. Pan, T.E. Miller, J.D. Potter, Isolation, expression and mutation of a rabbit skeletal

muscle cDNA clone for troponin I. The role of the NH2 terminus of fast skeletal muscle troponin I

in its biological activity, J. Biol. Chem. 267 (1992) 25407-25413.

[8] P. Cummins, S.V. Perry, Troponin I from human skeletal and cardiac muscles, Biochem. J. 171

(1978) 251-259.

[9] J. Maira, N. Genserb, D. Morandella, J. Maierb, P. Mairc, P. Lechleitnerb, C. Calzolarie, C. Laruee,

E. Ambachd, F. Dienstlb, B. Pauf, B. Puschendorfa, Cardiac troponin I in the diagnosis of

myocardial injury and infraction, Clin. Chim. Acta 245 (1996) 19-38.

[10] S.V. Perry, Troponin I: Inhibitor or facilitator, Mol. Cell. Biochem. 190 (1999) 9-32.

[11] B.H. Syska, J.M. Wilkinson, R.J.A. Grand, S.V. Perry, The relationship between biological activity

and primary structure of troponin I from white skeletal muscle of the rabbit, Biochem. J. 153

(1976) 375-387.

[12] S.V. Perry, Troponin T: Genetics, properties and function, J. Muscle Res. Cell Motil. 19 (1998)

575-602.

[13] D.G. Ward, S.M. Brewer, M.P. Cornes, I.P. Trayer, A cross-linking study of the N-terminal

extension of human cardiac troponin I, Biochemistry 42 (2003) 10324-10332.

[14] R. Ruksana, K. Kuroda, H. Terami, T. Bando, S. Kitaoka, T. Takaya, Y. Sakube, H. Kagawa,

Tissue expression of four troponin I genes and their molecular interactions with two troponin C

- 18 -

isoforms in Caenorhabditis elegans, Genes to Cells 10 (2005) 261-276.

[15] H. Terami, B.D. Williams, S. Kitamura, Y. Sakube, S. Matsumoto, S. Doi, T. Obinata, H. Kagawa,

Genomic organization, expression and analysis of the troponin C gene pat-10 of Caenorhabditis

elegans, J. Cell Biol. 146 (1999) 193-202.

[16] H. Kagawa, K. Sugimoto, H. Masumoto, T. Inoue, H. Imadzu, K. Takuwa, Y. Sakube, Genome

structure, mapping and expression of the tropomyosin gene tmy-1 of Caenorhabditis elegans, J.

Mol. Biol. 251(1995) 603-613.

[17] A. Anyanful, Y. Sakube, K. Takuwa, H. Kagawa, The third and fourth tropomyosin isoforms of

Caenorhabditis elegans are expressed in the pharynx and intestines and are essential for

development and morphology, J. Mol. Biol. 313 (2001) 525-537.

[18] C.D. Myers, P. Y. Goh, T.S. Allen, E.A. Bucher, T. Bogaert, Developmental genetic analysis of

troponin T mutations in striated and nonstriated muscle cells of Caenorhabditis elegans, J. Cell

Biol. 132 (1996) 1061-1077.

[19] H. Tanaka, Y. Takeya, T. Doi, F. Yumoto, M. Tanokura, I. Ohtsuki, K. Nishita, T. Ojima,

Comparative studies on the functional roles of N-and C-terminal regions of molluskan and

vertebrate troponin I, FEBS J. 272 (2005) 4475-4486.

[20] E. al-Hillawi, S.D. Minchin, I.P. Trayer, Overexpression of human cardiac troponin-I and

troponin-C in Escherichia coli and their purification and characterisation. Two point mutations

allow high-level expression of troponin-I, Eur. J. Biochem. 225 (1994) 1195-1201.

[21] H. Kagawa, T. Takaya, R. Ruksana, F. Anokye-Danso, Md. Z. Amin, H. Terami, C. elegans model

for studying tropomyosin and troponin regulations of muscle contraction and animal behavior. In: S.

Ebashi, I. Ohtsuki (Eds.), Regulatory mechanisms of striated muscle contraction. Adv. Exp. Med.

and Biol., vol. 594, Springer, Tokyo, 2007, pp153-161.

[22] P.K. Jha, P.C. Leavis, S. Sarkar, Interaction of deletion mutants of troponins I and T:

COOH-terminal truncation of troponin T abolishes troponin I binding and reduces Ca2+ sensitivity

of the reconstituted regulatory system, Biochemistry 35 (1996) 16573-16580.

[23] P.C.S. Chong, R.S. Hodges, Photochemical cross-linking between rabbit skeletal troponin subunits.

Troponin I-troponin T interactions, J. Biol. Chem. 257 (1982) 11667-11672.

[24] S.E. Hitchcock-DeGregori, Study of the structure of troponin I by measuring the relative

reactivities of lysines with acetic anhydride, J. Biol. Chem. 257 (1982) 7372-7380.

[25] C.S. Farah, C.A. Miyamoto, C.H.I. Ramos, A.C.R. da Silva, R.B. Quaggio, K. Fujimori, L.B.

Smillie, F.C. Reinach, Structural and regulatory functions of the NH2- and COOH-terminal regions

of skeletal muscle troponin I, J. Biol. Chem. 269 (1994) 5230-5240.

- 19 -

[26] D. Fisher, G. Wang, L.S. Tobacman, NH2-terminal truncation of skeletal muscle troponin T does

not alter the Ca2+ sensitivity of thin filament assembly, J. Biol. Chem. 270 (1995) 25455-25460.

[27] J.R. Pearlstone, L.B. Smillie, Effects of troponin-I plus-C on the binding of troponin-T and its

fragments to alpha-tropomyosin. Ca2+ sensitivity and cooperativity, J. Biol. Chem. 258 (1983)

2534-2542.

[28] A.J.M. Walhout, M. Vidal, High-throughtput yeast two-hybrid assays for large-scale protein

interaction mapping, Methods 24 (2001) 297-306.

[29] A.C. Gavin, M. Bosche, R. Krause, P. Grandi, M. Marzioch, A. Bauer, J. Shultz, J.M. Rick, A.M.

Michon, C.M. Cruciat, M. Remor, C. Hofert, M. Shelder, M. Brajenovic, H. Ruffner, A. Merino, K.

Klein, M. Hudak, D. Dickson, T. Rudi, V. Gnau, A. Bauch, S. Bastuck, B. Huhse, C. Leutwein,

M.A. Heurtier, R.R. Copley, A. Edelmann, E. Querfurth, V. Rybin, G. Drewes, M. Raida, T.

Bouwmeester, P. Bork, B. Seraphin, B. Kuster, G. Neubauer, G. Superti-Furga, Functional

organization of the yeast proteome by systematic analysis of protein complexes, Nature 415 (2002)

141-147.

- 20 -

Figure legends

Fig. 1. Schematic structures of four TNI isoforms and their constructed fragments and summary of

their molecular interactions. The fragments containing the number of amino acids are shown in

suffix. For example, the N-terminal fragment of TNI-1 containing amino acid residues 1-126 was

TNI-11–126. Amino acid residue number follows to the previous report [14]. Highlighted boxes

represent each location of the TNC-binding, TNT-binding, actin/TNC-binding and the C-terminal

extension, respectively. More details for TNT-binding site are presented in Fig. 6a. The bar

represents 50 amino acid residues. The right panel summarizes the molecular interactions of the

constructed fragments of TNI isoforms with TNC, TNT and TM. BW, body wall; PHX, pharynx.

“+” and “-” indicate a positive or negative interaction, respectively.

Fig. 2. Specificity of the antiserum against TNI, TNC and TM of C. elegans. (a) Cross-reactivity of

four TNI isoforms and their fragments with anti-CeTNI-2 antibody in Western blot analysis. The

lanes contained the total protein of E. coli harboring the expression constructs containing the N-,

C-terminal fragments of four TNI isoforms, C-terminal extension-deleted fragment of body wall

TNI-2 and four full-length TNI isoforms (Table 1). (b) Cross-reactivity of four TNI isoforms with

anti-CeTNI-4 antibody [14]. (c) Cross-reactivity of two TNC isoforms with the anti-CeTNC-1 and

anti-CeTNC-2 antibodies [14]. (d) Cross-reactivity of the body wall tropomyosin TMI and the

pharyngeal tropomyosin TMIII isoforms with anti-CeTMIII antibody in Western blot analysis [17].

The positions are indicated by the molecular size of marker proteins in kilo Daltons (kDa). Lane B

contained the total proteins in E. coli BL21 (DE3).

Fig. 3. Protein overlay assay of the body wall TNC-1 and the pharyngeal TNC-2 isoforms with

both body wall TNI-2 and pharyngeal TNI-4 isoforms in the presence or absence of Ca2+. (a)

Protein overlay assay of the TNC-1 isoform with the N- and C-terminal fragments of both TNI-2

and TNI-4 isoforms in the presence of Ca2+ (upper panel) or absence of Ca2+ (lower panel). The

overlaid TNC-1 isoform was detected by using the anti-CeTNC-1 antibody. (b) A similar protein

overlay assay between the TNC-2 isoform and TNI isoforms or their fragments. Overlaid TNC-2

was detected by using the anti-CeTNC-2 antibody. + Ca2+ and - Ca2+ indicates the presence and

absence of Ca2+, respectively. (c) Summary of the interactions between the TNI and TNC isoforms.

It was noted that the body wall TNC-1 only interacted with the N- terminal containing TNI-21–117

fragment, C-terminal extension-deleted TNI-21–220 fragment and full-length TNI-2.

- 21 -

Fig. 4. Protein overlay assay between the body wall and pharyngeal TNI isoforms and both tissues

TNT isoforms. Panels (a)-(h) demonstrate the results of the protein overlay assays; (a) TNI-21–117,

(b) TNI-294–242, (c) TNI-2118–242, (d) TNI-21–220, (e) TNI-2, (f) TNI-41–166, (g) TNI-491–194 and (h)

TNI-4 isoforms with both the body wall and pharyngeal TNT isoforms. The arrowheads indicate a

52-kDa TNT-1 and a 45-kDa TNT-4. Asterisks indicate no interaction with respective TNT

isoforms. The overlaid TNI-21–117, TNI-294–242, TNI-2118–242, TNI-21–220 and TNI-2 fragments were

detected by using the anti-CeTNI-2 antibody, and the overlaid TNI-41–166, TNI-491–194 and TNI-4

fragments were detected by using the anti-CeTNI-4 antibody. (i) Summary of interactions between

the TNI and TNT isoforms.

Fig. 5. Protein overlay assay of body wall TMI and pharyngeal TMIII with both the body wall

TNI-2 and pharyngeal TNI-4 isoforms. (a) Protein overlay assay of the TMI isoform with the N-

and C-terminal fragments of both TNI-2 and TNI-4 isoforms and the C-terminal extension-deleted

fragment TNI-21–220. (b) A similar assay of the TMIII isoform with those of TNI isoforms.

Overlaid TMI and TMIII were detected by using the anti-CeTMIII antibody. (c) Summary of

interactions between TNI and TM isoforms obtained from the results of (a) and (b).

Fig. 6. Alignment of the TNT-binding site of the TNI isoforms of invertebrates and vertebrates and

the molecular interaction model. (a) Sequence comparison of the TNT-binding sites of four TNI

isoforms of worms and those of Akazara scallop, crayfish, Drosophila, Ciona heart and body wall,

rabbit cardiac, human cardiac, rabbit slow, human slow, rabbit fast and human fast skeletal muscles.

The shade indicates the TNT-binding site of the TNI-2 isoform. The box indicates the IT-arm of the

human cardiac TNI [4]. The black bars represent the (H2) TNI helix region [4] and TNT-binding

site, respectively. Numbers in the TNT-, TNC- and actin/TNC-binding sites represent percentage

homology. Data follows the previous report [14]. Hydrophobic amino acids locating at the ‘a’ and

‘d’ positions in the heptads repeat are located at the top of the sequence. The identity and similarity

of the amino acid residues are indicated by (*) and (:/.), respectively. (b) The molecular interaction

model to study the interactions between the TNI isoforms and those with the other thin filament

components TNC, TNT and TM. Step 1 represents the interaction of the pharyngeal TNI-4 isoform

with that of pharyngeal TNC-2, TNT-4 and TMIII; Step 2 represents the presumably gene

duplication of pharyngeal TNI-4 to the body wall TNI isoforms; Step 3 represents the interaction of

the body wall type TNI isoforms with other pharyngeal thin filament proteins (as mentioned in step

- 22 -

1); Step 4 represents presumably gene duplication with mutations of pharyngeal isoforms to their

corresponding body wall isoforms; Step 5 represents the interaction of the body wall troponin and

tropomyosin isoforms with the body wall TNI isoforms, respectively. The rectangles and circles

indicate the body wall and pharynx specific isoforms, respectively.

- 23 -

BW, body wall; PHX, pharynx; ND, not determined; amolecular size on SDS-PAGE; bRuksana et al.[14];

cTerami et al. [15]; dWormBase; eKagawa et al.[16]; fProducts are used for the production of antibody.

Table 1 Construction and expression of TNI, TNC, TNT and TM

pCTNI-11-126 TNI-11-126 126 23

pCTNI-1106-250

ND

TNI-1106-250

pCTNI-2 TNI-2f

pCTNI-21-117 TNI-21-117

pCTNI-294-242 TNI-294-242

pCTNI-21-220 TNI-21-220

pCTNI-3 TNI-3

pCTNI-31-125 TNI-31-125

pCTNI-3118-260 TNI-3118-260

pCTNI-4 TNI-4f PHXb

pCTNI-41-166 TNI-41-166

pCTNI-491-194 TNI-491-194

pCTNC-1 TNC-1f BWc

pCTNC-2 PHXc

pCTNT-1 TNT-1

pCTNT-4 TNT-4

pCTMI TMI BWe

pCTMIII TMIIIf PHXe

BWd

PHXd

144

242

117

148

220

260

125

142

194

166

103

161

160TNC-2f

405

347

284

256

22

50

18

20

29

52

24

22

48

27

17

24

20

52

45

40

38

ND

ND

BWb

ND

ND

BWb

ND

ND

ND

ND

pCTNI-1

EcoRI

XhoI

EcoRI

BamHI

XhoI-HindIII

BamHI

EcoRI-Sma I

EcoRI

BglII-XhoI

BamHI-Sma I

EcoRI

XhoI

EcoRI

Sal I-EcoRI

EcoRI

BamHI

Hind III

Pst I-Hind III

EcoRI-Sma I TNI-1 250 50 BWb

Plasmid Restriction sites Product Length Molecular sizea Expressed tissue(aa) (kDa)

pCTNI-2118-242 TNI-2118-242 124 20 ND

BamHI

- 24 -

(PHX)

TMIII

+ + + + + +

+ +

++

+ + + +++

+ + + +

+ ++

+ +

+

- -

- -

--

- -

------

-

-

- - --

- -

--

-

-

Summary of molecular interactions

(BW) (PHX) (BW) (PHX) (BW)

+ +

+ + + + + +

+ + + + - -

- - - - - -

++

TNC-1 TNC-2 TNT-1 TNT-4 TMI

- -

+ +

+ +- - --

Fig. 1 Amin et al.

Schematic structure of four TNI isoforms and constructed fragments

50 amino acid

TNI-2

TNI-3

TNI-4

TNI-1

126

TNI-11-126

1

2501

106

TNI-1106-250

250

2421

TNI-21-117

1 117

TNI-294-242

24294

242

TNI-2118 -242

118

220

TNI-21-220

1

2601

125

TNI-31-125

1

118 260

TNI-3118-260

1 194

166

TNI-41-166

1

91 194 TNI-491-194

TNC-

bind

ing

C-te

rmin

al e

xten

sion

Act

in /T

NC-

bind

ing

TNT-

bind

ing

94 117

- 25 -

Fig. 2 Amin et al.

(a)

(c)

anti-CeTnC-1

B TNC

-1

TNC

-2

TNC

-2

TNC

-1

BkDa

17

30

42

anti-CeTnC-2

17

30

66

B TNI-

11-1

26

kDa TNI-

1106

-250

TNI-

1

TNI-

21-1

17

TNI-

2118

-242

TNI-

2TN

I-31

-125

TNI-

3118

-260

TNI-

3

TNI-

4

TNI-

491-

194

TNI-

41-1

66

TNI-

21-2

20

TNI-

294-

242

anti-CeTNI-2

(b)

anti-CeTNI-4

B TNI-

1

TNI-

2

TNI-

3

TNI-

4

TNI-

1

TNI-

2

TNI-

3

TNI-

4

TMI

TMII

I

B

anti-CeTMIII

(d)

- 26 -

Fig. 3 Amin et al.

+ Ca2+

- Ca2+

66

66

30

17

17

30

TNC-1

kDa

(b)(a)

TNC-2

anti-C eTNC-2

TNI-

2 1-1

17

TNI-

2 94-

242

TNI-

2 118

-242

TNI-

2 1-2

20

TNI-

2

TNI-

4 1-1

66

TNI-

4 91-

194

TNI-

4

TNC

-2

anti-CeTNC-1

TNI-

2 1-1

17

TNI-

2 94-

242

TNI-

2 118

-242

TNI-

2 1-2

20

TNI-

2

TNI-

4 1-1

66

TNI-

4 91-

194

TNI-

4

TNC

-1

anti-TNC-1

(c)

TNI-21-117

TNI-2

TNI-41-166

TNI-4

Body wall

Pharynx

Body wallTNC-1

Pharynx

TNC-2

TNI-21-220

- 27 -

Fig. 4 Amin et al.

(d)

kDa

(a) (c)(b)

TNT-

1

TNT-

4

TNT-

4

TNT-

1

TNT-

4*

TNT-

1*52

45

TNT-

4

TNT-

1*

TNT-

1*

TNT-

4

TNT-

4

TNT-

1*

anti-CeTNI-4

kDa45

(h)

TNI-2

TNI-4

Body wall

(e)

(f) (g)

TNI-2

TNI-21-117 TNI-2118-242 TNI-21-220

TNI-41-166 TNI-491-194

TNI-4

TNI-

4

TNI-41-166

TNI-21-117

TNI-21-220

TNI-491-194

TNI-

2 1-1

17

TNI-

2 118

-242

TNI-

2 1-2

20

TNI-

4 1-1

66

TNI-

4 91-

194 Body wall

TNT-1

Pharynx PharynxTNT-4

TNI-

2

TNT-

1

TNT-

4

anti-CeTNI-2

TNI-294-242

(i)

TNT-

4

TNT-

1

TNI-

2 94-

242

TNI-294-242

- 28 -

Fig. 5 Amin et al.

(c)

TN I-4

P hary nx

TN I-2

B ody w all

P hary nx

B ody w all

TM I

TM III

TNI-21-220

(a) (b)

TMIII

TNI-

2 1-1

17

TNI-

2 118

-242

TNI-

2 94-

242

TNI-

2 1-2

20

TNI-

2TN

I-4 1

-166

TNI-

4 91-

194

TNI-

4

TMIII

17

66

TMI

30

kD a TNI-

2 1-1

17

TNI-

2 118

-242

TNI-

2 94-

242

TNI-

2 1-2

20

TNI-

2TN

I-4 1

-166

TNI-

4 91-

194

TNI-

4

TMI

anti-CeTMIII

- 29 -

Fig. 6 Amin et al.

(b)

Body wall

TNC-1 TNT-1, - 2, - 3 TMI/II

Body wall

1

342

TNC-2 TNT-4 TMIII

5

Pharynx Pharynx

TNI-4

TNI-1, - 2, - 3

(a)

H2(TNI) helix region Identity(%) of Sequence name TNT-binding site the binding sites a d a d a d a d a d a d a d TNT TNC Actin/TNC CeTNI-1 BW 92 TVALPNVDSIDDHAKLEAIYNDLFSRLCNLEEEKYDINHITTETETTINQLNIEVNDLR 150 100 100 100 CeTNI-2 BW 80 TISLPNVDSIDDKGQLEKIYNDLWARLTQLEEEKYDINYVVSQTEAEINSLTIEVNDLR 138 72 81 100 CeTNI-3 BW 92 TIPLPDVDSINDQGQLLKIYEDMFARVCALEEEKFDINFGVSQTEAEINQLTIQVNDLR 150 59 71 94 CeTNI-4 PHX 79 IIPLPDLDNEDD---LEAVYDEIRERLIDLESENYDVSYIVRQKDFEINELTIAVNDLR 134 48 75 94 Akazara scallop 183 VPKFS--TDGKDVAALQALCKDFHKRLASLEEDVYDWEAKIRKQDFEINELTLKVNDTK 239 41 40 65 Crayfish 80 CGQPKNLDGANEE-QLRAIIKEYFDHTAQIESDKYDVELEIIRKDYEINELNIQVNDLR 137 31 60 88 Drosophila 78 CGSPRNLSDASED-TLKSLIKQHYDRINKLEDQKYDLEYVVKRKDVEINDLNAQVNDLR 135 24 60 94 Ciona heart 96 LEPLSG-LNSMSSQELMDLCRELHGKIDKVDEQRFDIEARVKKNDTEIEELNQKIFDLR 153 14 31 47 Ciona BW 49 LEPLSG-LNSMSSQELMDLCRELHGKIDKVDEQRFDIEARVKKNDTEIEELNQKIFDLR 106 14 27 53 Rabbit cardiac 80 CQPLE--LAGLGFAELQDLCRQLHARVDKVDEERYDVEAKVTKNITEIADLTQKIFDLR 136 28 19 53 Human cardiac 80 CQPLE--LAGLGFAELQDLCRQLHARVDKVDEERYDIEAKVTKNITEIADLTQKIFDLR 136 28 27 47 Rabbit slow 46 IPALQ--TRGLSLSALQDLCRQLHAKVEVVDEERYDIEAKCLHNTREIKDLKLKVLDLR 102 21 25 53 Human slow 49 IPTLQ--TRGLSLSALQDLCRELHAKVEVVDEERYDIEAKCLHNTREIKDLKLKVMDLR 105 21 31 53 Rabbit fast 50 CPPLS--LPGS-MAEVQELCKQLHAKIDAAEEEKYDMEIKVQKSSKELEDMNQKLFDLR 105 25 23 53 Human fast 50 CPPLH--IPGS-MSEVQELCKQLHAKIDAAEEEKYDMEVRVQKTSKELEDMNQKLFDLR 105 25 22 53 : : : : :.: :* . . : .:. : * :